High resolution radar example can not run

Omniverse Isaac Sim
, , ,
1

Isaac Sim Version

4.5.0

Operating System

Ubuntu 24.04

GPU Information

  • Model: RTX 4080 Super
  • Driver Version: 570

Running rtx radar example with high resolution config leads to crash.

Type below command
./python.sh /standalone_examples/api/isaacsim.util.debug_draw/rtx_radar.py -c 'Example_highRes'

ends up with this error:
2025-04-02 16:58:08 [36,895ms] [Error] [omni.sensors.nv.radar.wpm_dmatapprox.plugin] CUDA error 2 in ../../../source/extensions/omni.sensors.nv.common/include/omni/sensors/cuda/CudaHelperMem.h:56:out of memory (cudaMalloc(ptr, numElems * sizeof(T)))

We can not debug this cuda problem because the plugin is in .so.

Thanks.

2

You may run out of your GPU VRAM. Please monitor its usage using nvidia-smi.

3

Hi. Thanks for your response,

I have used an another GPU RTX 3090 which has 20GB comparing to my previous 4080 with 12 GB.

image
image720×405 105 KB

Now I can get the example running.

My ultimate goal is to get point cloud of ray cast that goes after one ray bouncing.

I know this type of ray tracing in only implemented in omni.sensors.nv.radar or USS with parameters
"TraceTreeDepth" : how many times the rays are reflected.

Are there other more straightforward approaches if I were to get point cloud after one ray bouncing?

Any suggestions are appreciated.

4

Could you create a new topic for this to keep discussions organized? Thanks.

1 Like